Towards Experiment Execution in Support of Community Benchmark Workflows for HPC

TL;DR

This paper introduces workflow templates for HPC benchmarks, enhancing experiment execution and adaptability across scientific applications, validated through two independent tools and diverse use cases.

Contribution

It proposes adaptable workflow templates for HPC benchmarks, validated by two tools, improving experiment management and supporting community-driven workflows.

Findings

Workflow templates improve experiment adaptability in HPC.

Validated tools support diverse scientific applications.

Focus on simple experiment management enhances educational use.

Abstract

A key hurdle is demonstrating compute resource capability with limited benchmarks. We propose workflow templates as a solution, offering adaptable designs for specific scientific applications. Our paper identifies common usage patterns for these templates, drawn from decades of HPC experience, including recent work with the MLCommons Science working group. We found that focusing on simple experiment management tools within the broader computational workflow improves adaptability, especially in education. This concept, which we term benchmark carpentry, is validated by two independent tools: Cloudmesh's Experiment Executor and Hewlett Packard Enterprise's SmartSim. Both frameworks, with significant functional overlap, have been tested across various scientific applications, including conduction cloudmask, earthquake prediction, simulation-AI/ML interactions, and the development of computational fluid dynamics surrogates.